Methodology for enhanced voice search experience

Abstract

Arrangements are described for reducing response latency in intelligent personal assistant applications. While receiving a user request, preemptive responses are automatically prepared for a received portion of the user request. Partial classification word candidates are generated for words in the received portion of the user request, and then a predictive component is applied to generate extended classification word candidates that include the partial classification word candidates and additional classification word candidates. A preliminary search is performed of the extended classification word candidates to prepare the preemptive responses. While the input request continues, the preemptive responses are updated, and when the input request ends, the prepared preemptive responses are used to respond to the user request.

What is claimed: 
     
       1. A method for an intelligent personal assistant application using at least one hardware implemented computer processor to respond to a user request containing a plurality of words, the method comprising:
 while receiving the user request:
 automatically preparing preemptive responses for a received portion of the user request based on:
 i. generating one or more partial classification word candidates for words in the received portion of the user request, 
 ii. applying a predictive component to the partial classification word candidates to generate one or more extended classification word candidates that include the partial classification word candidates and additional classification word candidates, and 
 iii. performing a preliminary search of the extended classification word candidates to prepare the preemptive responses; and 
 
 while the input request continues, continuing the preemptive response preparing and updating the preemptive responses; and 
 
 when the input request ends, using the prepared preemptive responses to respond to the user request. 
 
     
     
       2. The method according to  claim 1 , wherein the predictive component is based on a search language model. 
     
     
       3. The method according to  claim 1 , wherein the user request is a spoken user request. 
     
     
       4. The method according to  claim 3 , wherein automatic speech recognition of the spoken user request is used for generating one or more partial classification word candidates. 
     
     
       5. The method according to  claim 1 , wherein the user request is received from a remote device by a local server preparing the preemptive responses. 
     
     
       6. The method according to  claim 1 , wherein the preemptive responses reflect geographical constraints. 
     
     
       7. The method according to  claim 1 , wherein the preemptive responses reflect temporal constraints. 
     
     
       8. A system comprising:
 at least one processor; and 
 a memory device operatively connected to the at least one processor; 
 wherein, responsive to execution of program instructions accessible to the at least one processor, the at least one processor is configured to: 
 while receiving the user request:
 automatically prepare preemptive responses for a received portion of the user request based on:
 i. generating one or more partial classification word candidates for words in the received portion of the user request, 
 ii. applying a predictive component to the partial classification word candidates to generate one or more extended classification word candidates that include the partial classification word candidates and additional classification word candidates, and 
 iii. performing a preliminary search of the extended classification word candidates to prepare the preemptive responses; and 
 
 while the input request continues, continue the preemptive response preparing and updating the preemptive responses; and 
 
 when the input request ends, use the prepared preemptive responses to respond to the user request. 
 
     
     
       9. The system according to  claim 8 , wherein the predictive component is based on a search language model. 
     
     
       10. The system according to  claim 8 , wherein the user request is a spoken user request. 
     
     
       11. The system according to  claim 10 , wherein automatic speech recognition of the spoken user request is used for generating the one or more partial classification word candidates. 
     
     
       12. The system according to  claim 8 , wherein the user request is received from a remote device by a local server preparing the preemptive responses. 
     
     
       13. The system according to  claim 8 , wherein the preemptive responses reflect geographical constraints. 
     
     
       14. The system according to  claim 8 , wherein the preemptive responses reflect temporal constraints. 
     
     
       15. A computer program product encoded in a non-transitory computer-readable medium, which when executed by a computer causes the computer to perform the following operations:
 while receiving the user request:
 automatically preparing preemptive responses for a received portion of the user request based on:
 i. generating one or more partial classification word candidates for words in the received portion of the user request, 
 ii. applying a predictive component to the partial classification word candidates to generate one or more extended classification word candidates that include the partial classification word candidates and additional classification word candidates, and 
 iii. performing a preliminary search of the extended classification word candidates to prepare the preemptive responses; and 
 
 while the input request continues, continuing the preemptive response preparing and updating the preemptive responses; and 
 
 when the input request ends, using the prepared preemptive responses to respond to the user request. 
 
     
     
       16. The computer program product according to  claim 15 , wherein the predictive component is based on a search language model. 
     
     
       17. The computer program product according to  claim 15 , wherein the user request is a spoken user request. 
     
     
       18. The computer program product according to  claim 17 , wherein automatic speech recognition of the spoken user request is used for generating the one or more partial classification word candidates. 
     
     
       19. The computer program product according to  claim 15 , wherein the user request is received from a remote device by a local server preparing the preemptive responses. 
     
     
       20. The computer program product according to  claim 15 , wherein the preemptive responses reflect geographical constraints. 
     
     
       21. The computer program product according to  claim 15 , wherein the preemptive responses reflect temporal constraints.